Pulmonary hypertension manifests as an idiopathic form and as a complication of other circulatory conditions related to increases in pulmonary vascular resistance, increases in pulmonary arterial blood flow, and/or elevated left heart filling pressures. Five major pulmonary hypertension categories are recognized and organized into World Health Organization (WHO) groups based on similarities in clinical features and pathophysiologic and histologic findings ( Table 58-1 ). One of these categories, pulmonary hypertension due to lung diseases and/or hypoxemia (group 3) is discussed in detail in Chapter 59 and chronic thromboembolic pulmonary hypertension (group 4) is discussed in detail in Chapter 57 . This chapter focuses particularly on pulmonary arterial hypertension (PAH) and other conditions in group 1; it is primarily this group for which disease-specific therapies exist.
|GROUP 1: PULMONARY ARTERIAL HYPERTENSION (PAH)|
|GROUP 1′: PULMONARY VENO-OCCLUSIVE DISEASE (PVOD) AND/OR PULMONARY CAPILLARY HEMANGIOMATOSIS (PCH)|
|GROUP 1″: PERSISTENT PULMONARY HYPERTENSION OF THE NEWBORN|
|GROUP 2: PULMONARY HYPERTENSION DUE TO LEFT HEART DISEASE|
|GROUP 3: PULMONARY HYPERTENSION DUE TO LUNG DISEASES AND/OR HYPOXIA|
|GROUP 4: CHRONIC THROMBOEMBOLIC PULMONARY HYPERTENSION (CTEPH)|
|GROUP 5: PULMONARY HYPERTENSION WITH UNCLEAR MULTIFACTORIAL MECHANISMS|
Pulmonary hypertension in general is not uncommon; it is diagnosed in more than 2% of all patients discharged from U.S. hospitals and in up to 9% of echocardiograms performed in a community setting. The majority of pulmonary hypertension diagnoses relate to left heart disease or lung disease, with only a small fraction accounted for by PAH (group 1) and chronic thromboembolic pulmonary hypertension (group 4). Idiopathic PAH in particular is rare, with an estimated incidence of approximately 1 case per million and a prevalence of 7 cases per million. Because current disease-specific pulmonary hypertension medications are approved only for idiopathic PAH and other WHO group 1 conditions, as listed in Table 58-1 , it is critical for treating physicians to have a thorough understanding of the differential diagnosis and workup that is required for patients with suspected pulmonary hypertension.
What is now called idiopathic PAH was first described at autopsy by Romberg in 1891. Patients with pulmonary hypertension of unknown cause were reported infrequently over the next 60 years, but began to be recognized more commonly after the advent of cardiac catheterization in the 1940s.
Idiopathic PAH was initially described as a disease of young women. The National Institutes of Health registry, conducted from 1981 to 1988, reported a mean age of 36 ± 15 years with a female-to-male ratio of 1.7 : 1, and other registries—including the more recent REVEAL registry—have reported even higher ratios of female to male patients. However, patients of both genders and of all ages are affected, and older patients are now recognized more frequently, with a mean age of approximately 50 years reported in registries from both the United States and Europe.
Patients older than 65 years with idiopathic PAH are also being recognized with greater frequency in the current era than in the past. Although most of these patients have pulmonary hypertension secondary to left heart disease or lung disease, idiopathic PAH does present in this age group. Patients with idiopathic PAH in the older age range have a more balanced gender ratio (female-to-male 1.2 : 1), greater abnormalities in baseline functional class and 6-minute walk distance (6MWD), and a worse prognosis, despite having a lower pulmonary vascular resistance at diagnosis (average 8.3 Wood units vs. 12 Wood units in younger patients).
Wood, in 1958, divided pulmonary hypertension into six types: (1) passive, as seen with increased pulmonary venous pressure due to raised left atrial or ventricular pressure; (2) hyperkinetic, caused by increased pulmonary blood flow; (3) obstructive, resulting from pulmonary embolism or thrombosis; (4) obliterative, manifested by a reduction of pulmonary vascular capacity; (5) vasoconstrictive, brought about by functional and presumably reversible vasospasm; and (6) polygenic, arising in two or more of the preceding ways. Vasoconstrictive pulmonary hypertension was noted most reliably in association with acute alveolar hypoxia, but was also seen as an added component in some patients with other forms of pulmonary hypertension. Wood also hypothesized that widespread “obliterative” pulmonary vascular disease potentially might complicate virtually all varieties of long-standing, severe pulmonary hypertension. Because multiple mechanisms are now known to be common in severe pulmonary hypertension, regardless of underlying etiology, the modern classification has moved toward a more disease-focused classification. Nevertheless, these six pathophysiologic mechanisms remain relevant to modern understanding of the disease.
Also in 1958, Heath and Edwards documented the pathology of hypertensive pulmonary vascular disease in a study of 67 patients with congenital heart disease and 2 patients with idiopathic PAH, focusing on the muscular pulmonary arteries 100 to 1000 µm in size. These investigators argued that the progression of lesions in these patients was so stereotyped as to allow division of the structural effects of pulmonary hypertension into six grades as follows: grade 1, medial hypertrophy of the pulmonary arteries and arterioles without intimal changes; grade 2, medial hypertrophy with cellular intimal proliferation; grade 3, medial hypertrophy, intimal proliferation, and intimal fibrosis; grade 4, progressive generalized vascular dilation and occlusion by intimal fibrosis and fibroelastosis; grade 5, appearance of dilation lesions, including veinlike branches of occluded pulmonary arteries, plexiform lesions, angiomatoid lesions, and cavernous lesions; and grade 6, necrotizing arteritis.
Heath and Edwards lumped the pathology of PAH associated with congenital heart disease and idiopathic PAH together, and large series of patients with the latter disorder were not available until 1970. In that year, Wagenvoort and Wagenvoort described the morphology of the pulmonary vessels of 150 persons in whom a diagnosis of unexplained PAH had been made. The largest subset ( N = 110, 73%) had a histology consistent with what is now termed idiopathic PAH, and the pathologic abnormalities were strikingly similar to those described earlier by Heath and Edwards and Wood. The earliest abnormalities were medial hypertrophy of the muscular pulmonary arteries and muscularization of the arterioles, apparent in both children and adults with the disease. Less marked in children but apparent in all adults studied were intimal proliferation and laminar intimal fibrosis that gave an onionskin appearance to the pulmonary arteries, and plexiform lesions were found in 70%. A typical arteriopathic muscular artery is shown in Figure 58-1 .
A pathologic pattern consistent with chronic pulmonary thromboembolism was also seen in some of the patients with unexplained pulmonary hypertension in this series, present in 31 patients (21%). Half were noted in retrospect to have had other evidence of chronic thromboembolism, such as deep venous thrombosis. These patients had a combination of abnormalities including evidence of recent pulmonary thrombi, thrombi in the process of reorganization, and older organized fibrosis, appearing as intimal fibrosis, and lacked onion skin lesions, plexiform and other dilation lesions. Notably, this chronic thromboembolic pattern is rarely seen in patients diagnosed as idiopathic PAH in modern series, likely due to universal screening for chronic thromboembolic disease. These earlier histopathologic studies were, no doubt, limited by incomplete or erroneous phenotyping, in that the appropriate workup of pulmonary hypertension patients has only been clarified in the “recent” era, that is, when approved therapies became available.
Pulmonary veno-occlusive disease (PVOD) is a rare form of PAH that has virtually identical hemodynamic findings and in many cases similar clinical findings as in idiopathic PAH. It was seen in 3% of the pathology specimens in the series by Wagenvoort and Wagenvoort, and continues to be difficult, in many cases, to distinguish from idiopathic PAH using clinical means. The characteristic histologic feature of PVOD is obstruction of pulmonary venules and veins by loose, fibrous remodeling of the intima that may totally occlude the lumen; in addition arterial changes are usually evident.
Despite the distinct pathologic findings in idiopathic PAH, some degree of vascular remodeling takes place with essentially all forms of pulmonary hypertension, and even severe arteriopathic changes are, in fact, not unique to idiopathic PAH, having been reported in other forms of PAH (group 1) and in chronic thromboembolic pulmonary hypertension. For this reason, in addition to the risk of lung biopsy in patients with pulmonary hypertension, one rarely obtains pathologic tissue before lung transplantation or death. Similarly, lung biopsy in pulmonary hypertension due to congenital heart disease with left-to-right shunting is also no longer common.
Pathogenesis and Etiology
A dramatic change has taken place over the past several years in our thinking regarding the pathogenesis of idiopathic PAH. The “paradigm” has shifted from one of vasoconstriction to one of growth and proliferation. There are several lines of evidence suggesting that idiopathic PAH develops as a result of abnormal proliferation of vascular smooth muscle cells affecting all three layers of the vessel wall and leading to intimal hyperplasia, medial hypertrophy, and adventitial proliferation. What initiates this abnormal growth is not entirely known, but there are several clues. The concept of genetic predisposition toward growth and proliferation has more recently emerged. Mutations in the bone morphogenic protein receptor 2 (BMPR2) gene have been reported in patients with the familial form of idiopathic PAH. This gene contributes to the apoptotic process through a complex series of messenger proteins, as part of the transforming growth factor- β family of genes.
Accordingly, the possibility emerges that pulmonary arteriopathy is a failure of normal apoptosis. Some investigators have even referred to idiopathic PAH as “cancer of the pulmonary artery.” Although this is an attractive hypothesis, the story is not that simple. For one, the presence of a BMPR2 mutation is not always associated with the development of idiopathic PAH. It is likely that another genetic or acquired insult is required to initiate the arteriopathic process. Defects in a specific voltage-gated potassium ion channel, the Kv1.5 channel, have been found in the pulmonary artery smooth muscle cells from patients with idiopathic PAH. A defect or deficiency of this channel allows excess calcium to enter the cell and thus promotes both cell contraction and growth. Overexpression of the serotonin transporter has been described in patients with idiopathic PAH. This genetic defect might lead to increased internalization of serotonin and subsequent smooth muscle cell growth. Recent interest has focused on the role of impaired apoptosis on the development of proliferative arteriopathy in idiopathic PAH. It is clear that functional mutations seen in the BMPR2 gene with resultant abnormal downstream signaling results in impaired apoptosis.
In addition to potential genetic defects, several abnormalities in endothelial cell function have been found, many of which are likely a result of some vascular insult. The basis for currently approved therapeutics is the deficiency in prostacyclin and nitric oxide release and the excess in endothelin-1 and its receptor expression in patients with pulmonary hypertension. Some of the complex cellular and molecular processes contributing to idiopathic PAH are shown in Figure 58-2 .
Other Group 1 Conditions
Pulmonary Veno-occlusive Disease
Veno-occlusive disease has been associated with viral syndromes, toxin exposure, and chemotherapy, but may also be seen in an idiopathic form. Unlike idiopathic PAH, the male-to-female ratio in adults with PVOD is close to 1 : 1. A familial form of veno-occlusive disease also has been noted, including a subset of patients who carry the BMPR2 mutation. Differentiation from idiopathic PAH can be difficult. Suggestive findings include more hypoxia, a lower D l CO on pulmonary function testing, and prominent interlobular septal thickening (Kerley B) lines on chest radiographs in the absence of other signs of left-sided heart failure ( Fig. 58-3A ). Computed tomography (CT) scanning may reveal interlobular septal thickening (see Fig. 58-3B ), pleural effusions, and enlarged pulmonary arteries. The perfusion lung scan may show defects suggestive of thromboembolism but with unremarkable pulmonary angiograms. In addition, patients with PVOD may develop acute pulmonary edema in response to pharmacologic agents that reduce upstream pulmonary vascular resistance (PVR) and increase cardiac output, such as prostacyclin. This phenomenon probably is caused by an increase in pulmonary blood volume in the face of downstream vascular obstruction. A recent analysis of PVOD cases suggested that bronchoalveolar lavage (BAL) could be useful, because patients with PVOD almost always had increased hemosiderin-laden macrophages in BAL fluid.
Pulmonary Arterial Hypertension Associated with Other Conditions
Several systemic diseases or exposures have been shown, epidemiologically, to be associated with PAH (see Table 58-1 ). For example, approximately 8% of patients with the scleroderma spectrum of diseases, most notably limited scleroderma (previously CREST, or calcinosis cutis, Raynaud phenomenon, esophageal dysfunction, sclerodactyly, and telangiectasia) have been reported to have PAH ( eFig. 58-1 ) confirmed by right heart catheterization, and even higher rates have been seen in some echocardiography series. Pulmonary vascular involvement is associated with worse survival in scleroderma. Other connective tissue diseases associated with PAH, albeit less commonly, include systemic lupus erythematosus, mixed connective tissue disease, and rheumatoid arthritis.
Congenital heart disease is a well-recognized “risk factor” for development of PAH. In 1958, Wood coined the term “Eisenmenger’s complex” to describe pulmonary hypertension due to high PVR, with reversed (i.e., right-to-left) shunting through a large ventricular septal defect. Subsequently, the term has been used to describe pulmonary hypertension with cyanosis coupled with any systemic-to-pulmonary circulatory shunt. The likelihood of developing pulmonary hypertension depends on the size of the defect. However, patients with even small atrial septal defects can develop pulmonary hypertension. These patients may, in fact, have idiopathic PAH and the atrial septal defect merely serves as a “trigger” in a susceptible patient.
In patients with ventricular septal defects, it has been shown that only 3% of patients with a defect of 1.5 cm or smaller develop Eisenmenger syndrome, whereas 50% of patients with a large defect develop significant pulmonary hypertension, which also appears earlier than in atrial septal defect ( eFig. 58-2 ), often in infancy. PAH due to true Eisenmenger syndrome is associated with a longer natural history and more preserved right ventricular compensation and, hence, longer periods of clinical stability. The defect itself may also provide a protective mechanism, providing a route for blood to reach the underfilled left ventricle. However, right heart failure does eventually develop in many patients, and just as in idiopathic PAH, an elevated right atrial pressure and low cardiac output carry a poor prognosis.
The effects of drugs and toxins on the human pulmonary circulation have been graphically demonstrated in the past. Between 1967 and 1970 in Switzerland, Austria, and Germany, a 20-fold increase in pulmonary hypertension was observed after the introduction of aminorex, an appetite suppressant resembling epinephrine and amphetamine. The pulmonary lesions produced by this agent resembled plexogenic pulmonary arteriopathy in every respect. Pulmonary hypertension has also been associated with the appetite suppressants fenfluramine ( eFig. 58-3 ) and dexfenfluramine. In a case-controlled prospective study performed in Europe, the risk of developing pulmonary hypertension was increased by 20-fold in individuals who used one of these drugs for periods exceeding 3 months. Stimulants have also been suspected of causing pulmonary hypertension, based on both mechanistic similarities with the anorexigens and case reports, case series, and a single-center case-control study. In the latter study, an especially strong association between methamphetamine use and PAH was identified, with patients with PAH and no other identifiable risk factors 10 times more likely to have used methamphetamine than a matched control group of patients with other forms of PAH. Other medications suspected of causing pulmonary hypertension include chemotherapeutic agents, mainly in association with PVOD, and dasatinib, a tyrosine kinase inhibitor whose use in the treatment of chronic myelogenous leukemia has been associated with the development of PAH.
Portal hypertension is another disease associated with PAH ( eFig. 58-4 ) and is discussed further in Chapter 93 . Mantz and Craige first described simultaneous presentation of portal and pulmonary hypertension in a patient with portal vein thrombosis, speculating, as would others in the years following, that multiple emboli emanating from portacaval anastomoses were responsible for the pulmonary hypertension. Subsequent hypotheses have suggested a vasoactive mediator-based mechanism related to portosystemic shunting, whereby vasoactive substances, normally cleared in the liver, reach the pulmonary vasculature, while yet another hypothesis has focused on derangements in genes related to estrogen signaling. Portopulmonary hypertension is similar to other types of PAH histologically and is characterized hemodynamically by elevations in pulmonary arterial pressure and pulmonary vascular resistance. Incidence is estimated at 0.73% of all patients with portal hypertension, based on a large autopsy study, and rates of up to 4% to 5% have been reported in end-stage patients undergoing workup for liver transplantation.
In end-stage liver disease, hemodynamic assessment is frequently complicated by the presence of a high cardiac output related to splanchnic vasodilation and by the presence of volume overload and/or diastolic heart failure. As a result, portopulmonary hypertension patients tend to have higher cardiac outputs and lower PVRs compared with other WHO group 1 PAH patients. Despite more favorable hemodynamics, the mortality risk following liver transplantation has been shown to be approximately 50% in patients with even mild to moderate pulmonary hypertension (defined as a PVR >3.1 Wood units in this series), particularly when undiagnosed until the time of transplantation. Knowledge regarding the high mortality in patients with portopulmonary hypertension undergoing orthotopic liver transplantation has led to development of aggressive strategies for screening and treatment of these patients before transplantation; recommendations include echocardiography in all patients, right heart catheterization for patients whose estimated right ventricular systolic pressure exceeds 50 mm Hg, vasodilator treatment of patients whose pulmonary hypertension is due to PAH, and liver transplantation of these patients only when treatment is successful in lowering the mean PAH to less than approximately 35 mm Hg.
The association between human immunodeficiency virus (HIV) and PAH is well known. In several studies, the incidence of pulmonary hypertension in HIV has been reported as high as 0.5%. The mechanism of HIV-associated pulmonary hypertension is not known, although theories include release of cytokines and growth factors and the presence of human herpesvirus 8, a promoter of angiogenesis as seen in Kaposi sarcoma. One report noted the presence of human herpesvirus 8 in the cells from plexiform lesions in 10 of 16 patients with idiopathic PAH. Subsequent reports, however, failed to reveal human herpesvirus 8 in plexiform lesions. Pulmonary hypertension can develop in all stages of HIV, including in patients with no detectable viral load. A concomitant history of illicit intravenous drug use is noted in as many as 42% of cases, and at the authors’ Pulmonary Hypertension Center, the vast majority of patients with HIV-associated PAH have a methamphetamine use history, suggesting the possibility that illicit drugs contribute substantially to the risk of PAH in HIV patients. Clinical and hemodynamic features of HIV-associated pulmonary hypertension are similar to those of idiopathic PAH ( eFig. 58-5 ), although increased mortality is associated both with measures of HIV severity (high viral load, low CD4 count) and with typical PAH prognostic markers, such as worse functional class, low cardiac index, and history of right heart failure.
Dyspnea is the cardinal symptom of idiopathic PAH, described by more than 95% of patients in major clinical series, usually noted first on exertion and gradually with less and less activity. Closely related to dyspnea are sensations of fatigue and weakness, reported by a majority of patients with idiopathic PAH. These sensations usually are experienced before the general disability that is present with advanced disease, and presumably reflect impaired tissue oxygenation resulting from inadequate cardiac output.
Substernal chest pain is also commonly reported in patients with idiopathic PAH, and frequently arises on exertion, radiating to the left shoulder or axilla, and relieved by rest. Similarities with angina pectoris have led to the suggestion that the pain may relate to coronary insufficiency in the presence of increased right ventricular work and hypoxemia. However, pain may be present in young patients without coronary artery disease, and alternative mechanisms that have been considered include subendocardial ischemia related to demand-perfusion mismatch, ischemia related to direct compression of the left main coronary artery by the enlarged pulmonary trunk, and nonischemic pain caused directly by the distention of the pulmonary artery, whose afferents enter the nervous system along the same pathways as afferents from the heart. In addition to chest pain, hoarseness may result if the enlarged main pulmonary artery compresses the recurrent laryngeal nerve.
Syncope, usually with exertion, occurs in some patients with idiopathic PAH and may be its initial manifestation. Syncope is probably caused by a decrease in cerebral blood flow that follows an increase in pulmonary artery pressure and a decrease in cardiac output, and it is associated with a poor prognosis. Late findings in PAH include peripheral edema and ascites related to decompensated right ventricular failure, and hemoptysis. For the latter, bronchial artery embolization is often successful, but mortality following an initial episode of hemoptysis is high.
Patients with early idiopathic PAH may manifest no obvious physical abnormality. However, signs of pulmonary hypertension and right ventricular failure should be evident with advanced disease. As Wood observed, the hands and feet of a patient with severe pulmonary hypertension are cold, the peripheral pulse is diminished, the blood pressure is likely to be low, and the pulse pressure is reduced. Signs of systemic venous hypertension are often present, including a prominent jugular venous a wave, which is exaggerated by abdominal compression (hepatojugular reflux) and transmitted to the liver in a presystolic hepatic pulse, and prominent c-v waves, which are indicative of tricuspid regurgitation. Palpation of the chest may reveal a right ventricular lift at the left sternal border that is sustained throughout the pressure-overloaded cardiac contraction, in contrast to the unsustained parasternal impulse felt in pure volume overload.
On auscultation of the chest, the second heart sound is closely split, and the second (pulmonic) component is accentuated. The valvular closure sound should increase in intensity on inspiration and may become palpable as pulmonary artery pressure rises. A systolic ejection click reflecting sudden distention of the right ventricular wall also may be heard. A murmur of tricuspid regurgitation heard best along the left sternal border and increasing in intensity with inspiration, is frequent. A pulmonic regurgitant murmur may become evident after dilation of the main pulmonary artery and its valvular annulus. Diastolic vibration of the aortic valve leaflet (Graham Steell murmur) may be present along with third and fourth heart sounds. In addition to these findings, patients with right-sided heart failure usually have peripheral edema and abdominal distention due to ascites. If tricuspid regurgitation is present, the liver may become pulsatile.
Cyanosis is seen with variable frequency in patients with idiopathic PAH and is likely to be a late phenomenon. It is most marked during exercise but also may be present at rest. Peripheral vasoconstriction and impaired oxygenation of arterial blood due to mixed venous hypoxemia resulting from the decreased cardiac output appear to be the most common mechanisms. Patients in whom right atrial pressure equals or exceeds left atrial pressure may develop severe hypoxemia and cyanosis because of opening of the foramen ovale with subsequent right-to-left shunting of blood. In addition to cyanosis, vascular plethora may be observed in hypoxemic patients with secondary polycythemia. Clubbing is not a usual manifestation of idiopathic PAH. The presence of clubbing warrants a careful search for other causes of pulmonary vascular disease such as congenital heart disease, liver disease, PVOD, or idiopathic pulmonary fibrosis.
A thorough workup is required in all patients in whom PAH (WHO group 1) is suspected. A diagnostic algorithm has been developed, including basic tests—that should be completed in all patients—and optional tests that are performed when indicated ( Fig. 58-4 ). A similarly thorough workup will be required in many patients felt to have pulmonary hypertension related to a WHO group 2 or 3 condition, but a more limited evaluation may be appropriate in a subset of patients with advanced left heart or lung disease, borderline or mild pulmonary hypertension based on echocardiogram (right ventricular systolic pressure <50 mm Hg, no right ventricular dysfunction), and plans to focus therapy on the underlying disease process (i.e., no PAH-specific therapies).
The initial history should include a thorough review of symptoms as well as any potential pulmonary hypertension risk factors, including connective tissue disease, congenital heart disease, liver disease, obstructive sleep apnea, prior pulmonary embolism, and prior diet pill or stimulant use. Family history should be reviewed, including any family members with pulmonary hypertension, pulmonary embolism, or hereditary hemorrhagic telangiectasia. On physical examination, in addition to the potential physical findings described, particular attention should also be paid to findings that would suggest a non-idiopathic PAH diagnosis, such as significant crackles or wheezing on pulmonary examination, skin changes suggestive of connective tissue disease, pulmonary flow murmurs (suggestive of chronic thromboembolic disease), or evidence of clubbing.
Chest radiography is useful for suggesting the presence of pulmonary hypertension and in providing clues of underlying conditions such as parenchymal lung disease. In patients with idiopathic PAH, the radiograph characteristically reveals enlargement of the main pulmonary artery, increased width of the descending branch of the right pulmonary artery, peripheral oligemia, and an enlarged heart ( Fig. 58-5 and eFig. 58-6 ). The axial CT scan can confirm the increased diameter of the main pulmonary artery (see Fig. 58-5 ).
The electrocardiogram usually discloses right ventricular hypertrophy in patients with advanced idiopathic PAH. Electrocardiographic criteria for right ventricular hypertrophy include a QRS axis in the frontal plane that is greater than or equal to 110 degrees, an R wave in lead V1 that is greater than 5 mm, an R-to-S ratio in V1 that is greater than 1, and an R-to-S ratio in lead V6 that is less than 1. Patients also may manifest right atrial enlargement with a symmetrical and peaked P wave in lead II that is greater than 2.5 mm in amplitude. ST segment depression and T-wave inversion may be seen in the anterior chest leads. These abnormalities may not be present if pulmonary hypertension is not pronounced or if patients are young.
Systemic arterial blood gas analysis in patients with idiopathic PAH usually reveals a low arterial carbon dioxide pressure (arterial P co 2 ) and a normal pH, reflecting chronic respiratory alkalosis. The systemic arterial oxygen pressure (arterial P o 2 ) may be normal or abnormal, but the alveolar-to-arterial P o 2 difference usually is increased. Several mechanisms have been proposed for the hypoxemia of patients with idiopathic PAH, including diffusion impairment caused by a reduction in the number of pulmonary vessels coupled with the shortened time spent by erythrocytes in traversing the pulmonary circulation; ventilation-perfusion mismatching due to alterations in pulmonary blood flow; and concomitant conditions such as bronchospasm, right-to-left intracardiac shunting through a patent foramen ovale, and a reduced mixed venous P o 2 resulting from a low cardiac output. Blood studies are another important part of the laboratory evaluation. The complete blood count is particularly helpful in documenting polycythemia, which is present in hypoxemic patients with PAH or pulmonary hypertension related to lung disease.
Ventilation-perfusion lung scanning has been used primarily to differentiate idiopathic PAH from chronic pulmonary thromboembolism. Worsley and associates studied 75 patients with pulmonary hypertension of various types and found that 24 of 25 (96%) patients with chronic thromboembolic pulmonary hypertension had high probability scans, while one patient had an intermediate probability scan. In contrast, 33 of 35 patients (94%) with idiopathic PAH had a low probability scan ( eFig. 58-7 ), one had an intermediate probability scan, and one had a high probability scan. Based on these and other studies, patients with otherwise unexplained PAH who have an intermediate or high probability V/Q scan should undergo further evaluation, including pulmonary angiography. CT scanning is not recommended for evaluating chronic pulmonary embolism, because the sensitivity has been lower than V/Q scanning in head-to-head studies.
Pulmonary function tests performed in patients with idiopathic PAH usually reveal normal expiratory flow rates with normal or mildly reduced lung volumes. The modest restrictive defect has been attributed to diminished distensibility of the pulmonary vessels. The D l CO is often reduced to a mild or moderate degree, and a low D l CO out of proportion to lung volumes has been used in screening algorithms to identify scleroderma patients who are at increased risk of having pulmonary hypertension.
Exercise testing may serve to unmask physiologic abnormalities in patients with idiopathic PAH if these abnormalities are not present at rest. Characteristically, patients with PAH achieve their target heart rate and anaerobic threshold at low levels of exercise, often accompanied by a reduction in the arterial P o 2 or an increase in the alveolar-to-arterial P o 2 difference. The dead space–to–tidal volume ratio either fails to decrease, as it should in healthy persons, or actually increases during graded exercise. Submaximal exercise testing measured by the 6MWD is extremely useful as a prognostic marker and in following patients on therapy; individuals whose walk distance is greater than 380 to 400 m have a better prognosis. Additionally, patients with a postwalk test decline in heart rate of at least 16 beats/min have a better prognosis than those with a smaller decline, measured as heart rate at end of the test versus heart rate after 1 minute of recovery. The 6MWD has served as the primary outcome end point in several pivotal pulmonary hypertension clinical trials.
The echocardiogram is a key diagnostic test in the evaluation of PAH, providing an assessment of right ventricular size and function, an estimate of the pulmonary arterial systolic pressure, and in ruling out other cardiac conditions such as mitral valve disease and left ventricular systolic or diastolic dysfunction. Current guidelines recommend using echocardiography to estimate the likelihood of pulmonary hypertension, based on the Doppler-derived peak tricuspid jet velocity (TR jet) and using the equation: right ventricular systolic pressure = [(TR jet) × 4] + estimated central venous pressure. Specifically, a peak TR jet less than 2.8 m/sec is unlikely to represent PAH, whereas a velocity greater than 3.4 m/sec is consistent with probable pulmonary hypertension, and values between 2.8 and 3.4 m/sec, corresponding to estimated right ventricular systolic pressures of 37 to 50 mm Hg, are considered indeterminant.
Importantly, the accuracy and precision of this estimate compared with right heart catheterization measurement has been only moderate in many studies and, in certain circumstances, a decision to proceed with right heart catheterization may be appropriate even with a normal estimated pulmonary artery systolic pressure. It is also important to look for other echocardiographic findings suggestive of pulmonary hypertension. Suspicious findings include dilated right heart chambers, systolic flattening or diastolic bulging of the interventricular septum ( Fig. 58-6 and ), and a short pulmonary arterial acceleration time, although in general these tend to be later findings. Intracardiac shunting may be observed after the intravenous injection of contrast or microbubbles. Echocardiographic measurements may also be useful in estimating prognosis and tracking response to therapy. Importantly, this does not include estimated pulmonary arterial systolic pressure, which by itself does not have prognostic value in idiopathic PAH. Instead, echocardiographic measures suggesting right ventricular dysfunction should be sought, including dilated right heart chambers, reduced right ventricular systolic function, significant tricuspid regurgitation, marked septal shift with a small left ventricular chamber, and a pericardial effusion.
In the evaluation of idiopathic PAH, right heart catheterization is mandatory to document the presence and severity of pulmonary hypertension, rule out cardiac causes, and determine whether there is acute pulmonary vasoreactivity using pharmacologic agents. Hemodynamic values, especially right atrial pressure and cardiac index, correlate closely with survival. Cardiac chamber and pulmonary arterial pressures are recorded, and the pulmonary capillary wedge pressure (P pw ) is measured to rule out disease at the level of the left ventricle, left atrium, or large pulmonary veins. Cardiac output is measured, and the pulmonary and systemic vascular resistances are calculated. Blood gas samples are obtained to determine oxygen tensions and contents in the two circulations. Left-to-right intracardiac shunts may be excluded by the measurements of blood oxygen tensions and contents in the various cardiac chambers and by indicator techniques.
PAH—without left-sided heart involvement—is defined hemodynamically as a mean pulmonary arterial pressure greater than or equal to 25 mm Hg and a P pw less than 16 mm Hg. A PVR greater than or equal to 3 Wood units has been included in some but not other definitions, and the vast majority of patients with idiopathic PAH will easily meet this cutoff at diagnosis, with an average PVR at diagnosis of 12 Wood units or higher.
The measurement of P pw in pulmonary hypertension deserves extra attention, because a reading greater than 15 mm Hg suggests a diagnosis of left-sided heart disease rather than PAH. The P pw is obtained by transiently occluding blood flow in the pulmonary artery using an inflated, balloon-tipped catheter. The P pw can be inaccurate because of incomplete occlusion, resulting in a blunted pulmonary arterial pressure measurement rather than a true P pw , or because the catheter tip is not located centrally within the pulmonary artery. Inspecting the catheter location under fluoroscopy and ensuring that the resultant pressure tracings are consistent with a left atrial pressure waveform helps ensure an accurate reading. In some cases, a better waveform may be obtained by partially deflating the balloon and repositioning it. Additionally, a wedge position of the catheter can be confirmed by aspirating blood from the distal lumen and documenting high oxygen saturation indicative of pulmonary capillary blood. Because this measurement is so critical to the diagnosis, some centers routinely measure left ventricular end-diastolic pressures during all diagnostic right heart catheterizations.
Acute vasodilator testing is often performed during the initial catheterization. This study is performed using a short-acting agent, such as inhaled nitric oxide, adenosine, or prostacyclin. Oxygen and nitrates are not adequate testing agents in patients with idiopathic PAH. Criteria for defining vasoresponsiveness are discussed later, but basically, acute testing is performed to identify patients who will benefit from long-term vasodilator therapy.
Treatment and Prognosis
Idiopathic and other forms of PAH are now treatable diseases. Clear-cut short- and long-term benefits are seen with currently available therapies. To optimize a patient’s outcome, a comprehensive medical approach is essential. Once the diagnostic process in a patient with pulmonary hypertension is complete and the patient is characterized as having PAH (WHO group 1), therapy should be initiated. But many questions arise regarding PAH therapy. What are the goals and expected outcomes of treatment? Which drug should be used first? When should another therapy be added? Should more than one therapy be used? In what order? When should transplantation be considered?
Therapy for PAH may be subdivided into “supportive” or “conventional” therapies, defined as empirical treatments or recommendations for which there is no prospective, randomized, controlled data, and “specific” or “targeted” therapies, which have been tested and approved by regulatory authorities for the treatment of PAH.
Exercise and Physical Activity
Although data are limited, consensus guidelines support the benefits of exercise in PAH patients. However, patients should avoid activities that lead to undue symptoms such as severe dyspnea, chest pain, light-headedness, or syncope. Two small studies ( N = 22 and N = 30) have demonstrated that exercise and respiratory training can be safe and lead to measurable improvements in subjective and objective parameters. In these studies, patients undergoing 12- to 15-week courses of supervised aerobic exercise and resistance training were found to have significant improvement in 6MWD outcomes and peak oxygen consumption compared to controls. No significant safety concerns were identified, and there were no significant changes in echocardiographic measures of right heart function or in brain natriuretic peptide levels.
Avoidance of Altitude
Hypobaric hypoxia causes pulmonary vasoconstriction and, thus, can worsen pulmonary hypertension and lead to symptomatic worsening in PAH patients. It is generally recommended that patients flying on commercial airliners (pressurized to 1500 to 2400 m) or traveling to elevation above 5000 feet be evaluated for supplemental oxygen (see Chapter 25 ). Not surprisingly, patients with severe PAH residing at high elevations may improve if they move to sea level.
Avoidance of Pregnancy
Pregnancy is extremely risky in patients with PAH, with a high peripartum mortality rate, especially after delivery. Although there are case reports of patients managed with epoprostenol and undergoing successful pregnancies and deliveries, it is strongly recommended that women of childbearing potential use appropriate methods of birth control to avoid pregnancy. In terms of which birth control method is preferable, none of the highly effective methods are absolutely contraindicated in PAH, although surgical procedures (sterilization) may be too high risk for many patients. Additionally, some experts recommend avoiding estrogen-containing hormonal contraceptives, particularly in patients who are not anticoagulated. Efficacy can be generally thought of in three tiers: (1) sterilization, intrauterine devices, and progestin-containing implants; (2) combination and progestin-only birth control pills, the estrogen containing transvaginal ring, and injectable progestins; and (3) barrier methods, such as the condom and diaphragm, with the latter choices considered appropriate in PAH only when used in combination with another method.
There is strong rationale for the use of anticoagulants in PAH. Many of the endothelial cell abnormalities that predispose patients to pulmonary arteriopathy also increase thrombosis. The presence of heart failure and an indwelling central venous catheter are independent risk factors for thromboembolic events, which are poorly tolerated by patients with an already marginal pulmonary vascular reserve. In addition, microscopic thrombotic lesions in the pulmonary vasculature are well documented in PAH patients.
Warfarin is the anticoagulant most frequently used in patients with PAH. In PAH clinical trial registries, about 50% to 85% of patients are on anticoagulants at study entry. However, anticoagulation has risks (i.e., bleeding) as well as the need for frequent monitoring. To justify the use of warfarin in PAH, nine studies have examined the effects of warfarin in idiopathic PAH. All were retrospective analyses with some patients on warfarin and others untreated. Although better survival was documented in patients on warfarin compared with those not anticoagulated in six of nine studies, including the study by Rich and coworkers, as shown in Figure 58-7 , none of them were randomized and few were conducted in the modern era of effective PAH therapies. There have been two studies using warfarin conducted in the modern PAH therapy era; one resulted in better survival in idiopathic PAH, while the other was inconclusive.
Despite the serious limitations in the existing data, published guidelines recommend that patients with idiopathic PAH be treated with warfarin. There is less guidance regarding the use of anticoagulation in other forms of PAH, such as that associated with congenital systemic-to-pulmonary shunts or associated with connective tissue diseases. Warfarin is clearly indicated for patients with Group 4 PAH (chronic thromboembolic disease). Other potential situations in which warfarin may be considered include advanced heart failure or the presence of indwelling central venous catheters. Conversely, if there is increased risk for bleeding (thrombocytopenia, history of hemoptysis, or gastrointestinal bleeding), withholding anticoagulation is advised.
Factor Xa Inhibitors.
Despite its pharmacologic limitations and need for frequent coagulation monitoring, warfarin, since 1954, has had virtually no competition as the clinical agent of choice for long-term anticoagulation. Recently, however, a novel group of oral anticoagulants—known as factor Xa inhibitors—has altered the current anticoagulation picture. Most clinical studies have compared the efficacy and safety of the standard regimen of parenteral enoxaparin followed by warfarin compared with a fixed-dose of a factor Xa inhibitor for treatment of either acute venous thromboembolism or atrial fibrillation; results to date indicate that long-term warfarin has equal anticoagulant effectiveness as one of several factor Xa inhibitors. Furthermore, an experimental study of a factor Xa inhibitor using a rat model of PAH also showed beneficial results. No effective antidote for serious bleeding related to factor Xa inhibitors is yet available, but major hemorrhage is infrequent. More information about this promising group of compounds is eagerly awaited.
Aspirin has also been considered potentially beneficial in PAH, based on evidence of platelet activation in PAH and on evidence of improved outcomes in animal models of pulmonary hypertension. However, its use did not lead to improved outcomes in a randomized controlled clinical trial in PAH and it is therefore not recommended unless required for other indications.
The benefits of supplemental oxygen in PAH patients, unlike patients with pulmonary hypertension associated with lung diseases such as chronic obstructive pulmonary disease, are not clear. In fact, most PAH patients are not hypoxemic at rest. Mild hypoxemia, when present, is likely on the basis of reduced mixed venous oxygen saturation levels caused by low cardiac output with mild ventilation/perfusion inequality. The presence of more profound hypoxemia in a patient with PAH should raise suspicion for underlying parenchymal lung disease, systemic to pulmonary shunting, PVOD, pulmonary capillary hemangiomatosis, or pulmonary arteriovenous malformations as seen in pulmonary hypertension due to hereditary hemorrhagic telangiectasia. Of note, a patent foramen ovale, present in more than 20% of the population, can contribute to hypoxemia in pulmonary hypertension.
Although oxygen is a pulmonary vasodilator, there are no long-term studies supporting its efficacy. However, the consensus is that, if arterial P o 2 is less than 60 mm Hg or systemic arterial O 2 saturation is less than 90% at rest, supplemental oxygen is indicated. One exception to this approach is in patients with Eisenmenger syndrome, with hypoxemia due to right-to-left shunting; in this group, the use of supplemental oxygen may have negligible benefit. There is also no general agreement about whether exercise-only systemic arterial O 2 desaturation warrants oxygen supplementation. In addition, the “stigma” of nasal cannulae for a PAH patient often limits compliance outside the home.
Diuretics have long been mainstays of therapy for heart failure, including right ventricular failure. Both total-body and intravascular volume overload are common in PAH patients. In pivotal trials of PAH drugs, the majority of patients were on chronic diuretic therapy.
In addition to causing symptomatic peripheral edema and renal and hepatic congestion, volume overload of the right ventricle can cause compression of the left ventricle and contribute to decreased cardiac output and prerenal azotemia. Thus, it is a common observation that in decompensated PAH patients, aggressive diuresis leads to clinical and physiologic improvements. Despite the benefits of diuretics in PAH patients, there are no controlled studies to guide the clinician in using these agents.
A loop diuretic is frequently used first. Although furosemide is often the loop diuretic of choice, there is some evidence that torsemide might be more efficacious without increased side effects. In addition, in patients with marked extravascular fluid accumulation and poor intestinal absorption, intravenous diuretic therapy is frequently needed. Anti-aldosterone drugs (e.g., spironolactone) are commonly combined with loop diuretics in PAH patients. Whether data suggesting a morbidity and mortality benefit of spironolactone in left-sided heart failure can be extrapolated to right-sided heart failure is not known.
In some cases, the addition of a thiazide diuretic to the regimen is appropriate. The combination of metolazone and furosemide has been found by the authors to be effective in causing a brisk diuresis. However, marked hypokalemia can be seen with this regimen. Although it is possible that in some patients with PAH, the right ventricle is preload dependent and therefore over diuresis can be detrimental, this has not been the authors’ experience. More often, aggressive diuresis (1- to 3-L negative fluid balance per day) leads to improvement in renal function and blood pressure.
Calcium Channel Antagonists
In 1958, Paul Wood first defined the clinical entity of pulmonary hypertension with reference to the “vasoconstrictive factor.” It is not surprising, then, that a search for pulmonary vasodilators as effective therapies ensued. Agents including phentolamine, tolazine, captopril, and hydralazine were evaluated in uncontrolled reports. Results, although variable, were not overwhelmingly favorable. No systematic studies of these medications were carried out. Out of the myriad of oral antihypertensive agents emerged calcium channel blockers. Ostensibly, this class of agents “made sense” for treating pulmonary hypertension. Calcium channel blockers have acceptable side effect profiles and are potent pulmonary as well as systemic vasodilator agents. The role of intracellular cytosolic calcium in the vasoconstriction of pulmonary artery smooth cells was well established. Thus, blocking influx of calcium into the cells seemed desirable.
In a highly quoted paper, Rich and associates described favorable survival in a subgroup of idiopathic PAH patients treated with either diltiazem or nifedipine. In that study, patients manifesting acute pulmonary vasoreactivity with calcium channel blockers, defined as an acute decrease in mean pulmonary arterial pressure and PVR of at least 20%, had a 5-year survival of 94% (see Fig. 58-7 ). In contrast, patients who did not have an acute response had a 5-year survival of only 55%. In addition, the observed survival for the “acute responders” was significantly better than the survival predicted using an equation based on hemodynamics at time of diagnosis. Although not a placebo-controlled study, these data suggested a benefit with calcium channel blockers in some idiopathic PAH patients.
The Rich and associates’ study, although seminal, likely led to overuse of calcium channel blockers, not only for idiopathic PAH but for other forms of PAH. Calcium channel blockers are not selective pulmonary vasodilators and, in the setting of a nondilatable pulmonary vascular bed, the systemic vasodilating effects of these agents may lead to severe symptomatic hypotension. In addition, calcium channel blockers have potential negative inotropic effects. Thus, in patients with minimal or no acute pulmonary vasoreactivity, the negative effects of calcium channel blockers can become predominant, with potential for catastrophic consequences.
A subsequent large retrospective study by Sitbon and colleagues has further narrowed the role of calcium channel blockers in patients with idiopathic PAH. In that study, 557 patients with idiopathic PAH were tested for acute pulmonary vasoreactivity using either intravenous prostacyclin (e.g., epoprostenol) or inhaled nitric oxide. Seventy patients (about 13%) had an acute response (at least a 20% fall in mean pulmonary artery pressure and PVR) and were treated with calcium channel blockers. However, only half of those “acute responders” (about 7% of the total) did “well” long term on calcium channel blockers, defined as being alive and in functional class 1 or 2 at 5-year follow-up. The long-term survivors had less severe disease as assessed by hemodynamics at baseline and had reached a lower mean pulmonary artery pressure with acute vasodilator challenge than those who failed with the long-term calcium channel blockers (33 mm Hg vs. 46 mm Hg). In other words, rather than the percent drop in mean pulmonary artery pressure during an acute test, the absolute mean pulmonary artery pressure reached appeared to be better in defining the patients who would benefit long-term with calcium channel blockers. These data have been codified into evidence-based guidelines, which define potential calcium channel blocker candidates as idiopathic PAH patients in whom, during an acute pulmonary vasoreactivity test, the mean pulmonary artery pressure decreases by at least 10 mm Hg to a level below 40 mm Hg, with no decrease in cardiac output.
The method for performing acute pulmonary vasoreactivity testing varies among pulmonary hypertension centers. Most frequently, one of three short-acting pulmonary vasodilators is used in the cardiac catheterization laboratory (i.e., inhaled nitric oxide, intravenous adenosine, or intravenous epoprostenol). Using a short-acting agent prevents refractory systemic hypotension that could result when a patient with PAH with minimal pulmonary vasoreactivity is given a systemic vasodilator. A distinct advantage of inhaled nitric oxide is the absence of systemic hemodynamic effects, the very rapid “on-off” properties of the drug (i.e., half-life 20 seconds), and the absence of side effects. With inhaled nitric oxide, an acute pulmonary vasoreactivity test with repeat hemodynamic measurements can be accomplished in less than 10 minutes.
PAH-specific therapies have been available since 1995, when intravenous epoprostenol was approved by the U.S. Food and Drug Administration (FDA), based on the first prospective randomized controlled trial done in PAH. Since then, an additional 11 therapies have been approved for PAH: subcutaneous, inhaled, intravenous, and oral treprostinil; inhaled iloprost; the oral endothelin receptor antagonists bosentan, ambrisentan, and macitentan; the phosphodiesterase-5 inhibitors sildenafil and tadalafil; and the guanylate cyclase stimulator riociguat ( Table 58-2 ). These therapies are targeted to offset the imbalance in endothelial-derived mediators seen in PAH: excessive endothelin-1 production, abnormal nitric oxide production, and deficient prostacyclin ( Fig. 58-8 ).